Overall Accuracy For Detection Research Articles

Contrast-enhanced ultrasound with microbubbles containing sulfur hexafluoride and perfluorobutane with Kupffer phase for the detection of colorectal liver metastases.

To compare contrast-enhanced ultrasound (CEUS) with microbubbles containing sulfur hexafluoride (SHF) and perfluorobutane (PFB) for the detection of colorectal liver metastasis (CRLM). In this prospective study, conducted from September to November 2021, patients with colorectal cancer were consecutively recruited and underwent same-day ultrasound, SHF-CEUS, and PFB-CEUS. The reference standard was contrast-enhanced MRI and follow-up imaging. The size, depth, echogenicity, and calcification of each focal liver lesion were recorded. The number and conspicuity of CRLMs, based on washout appearance during the late phase (LP) (> 120 s)/Kupffer phase (KP), were evaluated offsite by two blinded readers. Overall, 230 lesions (CRLMs, n= 219; benign lesions, n= 11) in 78 patients were evaluated. Lesion conspicuity (p= 0.344) and accuracy in the detection of CRLM were comparable for SHF- and PFB-CEUS (0.877 for SHF vs. 0.770 for PFB, p= 0.087). More CRLMs ≥ 10 mm were identified by LP contrast washout in SHF-CEUS than in KP PFB-CEUS (p< 0.001). More CRLMs < 10 mm were identified by KP washout in PFB-CEUS than in LP SHF-CEUS (p< 0.001). Conspicuity was better on PFB-CEUS than on SHF-CEUS (p= 0.027). In hyperechoic lesions, lesions located deeper than 80 mm, and calcified lesions, CRLM conspicuity on PFB-CEUS was inferior to that on SHF-CEUS (p< 0.05). The overall accuracy of detection and conspicuity of washout in CRLMs were comparable between SHF and PFB-CEUS. PFB-CEUS has the advantage of identifying washout in small CRLMs. However, larger, hyperechogenic, deep-seated, or calcified lesions were better identified using SHF-CEUS. Accuracy of detection and conspicuity of washout in CRLMs were comparable between SHF- and PFB-CEUS. PFB-CEUS has the advantage in detecting small CRLMs, whereas SHF-CEUS is better for detecting larger, hyperechogenic, deep-seated, or calcified lesions. Contrast-enhanced ultrasound with sulfur hexafluoride in the late phase and perfluorobutane microbubbles in the Kupffer phase were comparable in terms of accuracy in the detection and conspicuity of colorectal liver metastases. Small colorectal liver metastases (< 10 mm) were more often identified in the Kupffer phase contrast-enhanced ultrasound imaging when using perfluorobutane microbubbles. Larger, hyperechogenic, deep-seated, or calcified lesions were better identified in the late phase contrast-enhanced ultrasound imaging (> 120 s) when using sulfur hexafluoride microbubbles.

Leveraging Saliency in Single-Stage Multi-Label Concrete Defect Detection Using Unmanned Aerial Vehicle Imagery

Visual inspection of concrete structures using Unmanned Areal Vehicle (UAV) imagery is a challenging task due to the variability of defects’ size and appearance. This paper proposes a high-performance model for automatic and fast detection of bridge concrete defects using UAV-acquired images. Our method, coined the Saliency-based Multi-label Defect Detector (SMDD-Net), combines pyramidal feature extraction and attention through a one-stage concrete defect detection model. The attention module extracts local and global saliency features, which are scaled and integrated with the pyramidal feature extraction module of the network using the max-pooling, multiplication, and residual skip connections operations. This has the effect of enhancing the localisation of small and low-contrast defects, as well as the overall accuracy of detection in varying image acquisition ranges. Finally, a multi-label loss function detection is used to identify and localise overlapping defects. The experimental results on a standard dataset and real-world images demonstrated the performance of SMDD-Net with regard to state-of-the-art techniques. The accuracy and computational efficiency of SMDD-Net make it a suitable method for UAV-based bridge structure inspection.

An Increased Diagnostic Accuracy of Significant Coronary Artery Stenosis Using 320-slice Computed Tomography with Model-based Iterative Reconstruction in Cases with Severely calcified Coronary Arteries.

Purpose High-quality images can be obtained with 320-slice computed tomography (CT) with model-based iterative reconstruction (MBIR). We therefore investigated the diagnostic accuracy of 320-slice CT with MBIR for detecting significant coronary artery stenosis. Methods This was a retrospective study of 160 patients who underwent coronary CT and invasive coronary angiography (ICA). The first 100 consecutive patients (Group 1) underwent 320-slice CT without MBIR or small-focus scanning. The next 60 consecutive patients (Group 2) underwent 320-slice CT with both MBIR and small-focus scanning. Patients who underwent coronary artery bypass surgery were excluded. The diagnostic performance of 320-slice CT without MBIR or small-focus scanning and 320-slice CT with both of them, with ICA regarded as a reference standard, was compared to detect significant coronary artery stenosis (≥70% on CT, ≥75% on ICA). Results In a patient-based analysis, the sensitivity, specificity, and overall accuracy of detection of significant stenosis on CT against ICA were 95%, 85%, and 91% in Group 1, and 93%, 83%, and 90% in Group 2, respectively. No significant differences were observed between the two groups in the patient- and segment-based analyses. However, among cases with a severe coronary artery calcium score >400 (31 cases in Group 1 and 28 in Group 2), the specificity and overall accuracy were significantly higher (all p<0.01) in Group 2 than in Group 1 according to the segment-based analysis. Conclusion The diagnostic accuracy of the detection of coronary artery stenosis on CT was improved using 320-slice CT with MBIR.

Diagnostic accuracy of PremaQuick in detection of preterm labor in symptomatic women.

Failure to identify women at risk of preterm labor (PTL) leads to failure to implement standard measures. This study designed to evaluate the accuracy of PremaQuick test in detection of PTL in women presented with threatened preterm labor (TPTL). One hundred and twenty-two (122) pregnant women, singleton pregnancy, < 37 weeks, admitted with TPTL included in this study, and were compared to 122 controls. After thorough evaluation, participants were examined using sterile vaginal speculum for cervico-vaginal fluid (CVF) sampling, and PremaQuick test. The CVF sampling was followed by trans-vaginal sonographic (TVS) assessment of cervical length (CL). Participants were managed according to hospitals policy thorough their admission, and after discharge in the ante-natal clinics till delivery. After delivery, the delivery data were compared by the recorded participants` data on admission. The PremaQuick test had 95.1% sensitivity, 97.5% specificity, 97.5% positive predictive value, 95.2% negative predictive value, and 96.3% accuracy in detection of PTL. The PremaQuick had significantly higher true negative rate, specificity, positive predictive value, and overall accuracy in detection of PTL compared to CL < 25 mm (p = 0.005, 0.005, 0.01, 0.002; respectively). The PremaQuick is an accurate bedside test in detection of PTL in women presented with TPTL. It had 95.1% sensitivity, 97.5% specificity, 97.5% positive predictive value, 95.2% negative predictive value, and 96.3% overall accuracy in detection of PTL. The PremaQuick had significantly higher true negative rate, specificity, positive predictive value, and overall accuracy in detection of PTL compared to CL < 25 mm.

Protein/creatinine ratio versus 24-hours urine protein in preeclampsia.

Proteinuria is one of the diagnostic criteria of preeclampsia (PE). Measurement of 24-hour urine protein is the gold standard method for detection of proteinuria in PE. The 24-hour urine sampling is time-consuming, and inconvenient. To evaluate the accuracy of protein/creatinine (P/C) ratio in detection of significant proteinuria (> 1 g/24-hours urine) in PE. One hundred and ten (110) preeclamptic women were included in this study and admitted for blood pressure monitoring, 24-hour urine collection, fetal well-being assessment and spot urine sample for measuring of P/C ratio. After thorough history and clinical examination, routine antenatal investigations were done for the women included in the study according to the hospital`s protocol, and to exclude pre-existing chronic renal disease. Twenty-four-hour urine collection started on the morning following hospital admission. Spot urine samples were obtained shortly before the 24-hour urine collection for measuring of P/C ratio. Collected data statistically analyzed to evaluate the accuracy of P/C ratio in detection of significant proteinuria in PE. The P/C ratio of 1.35 ± 2.54 had 94.4% sensitivity, 94.9% specificity, 97.1% positive predictive value (PPV), 90.2% negative predictive value (NPV), and 94.5% overall accuracy in detection of significant proteinuria (> 1 g/24-hour urine) in PE. The P/C ratio of 1.35 ± 2.54 had 94.4% sensitivity, 94.9% specificity, 97.1% PPV, 90.2% NPV, and 94.5% overall accuracy in detection of significant proteinuria (> 1 g/24-hour urine) in PE. This study suggests the use of P/C ratio as an alternative to 24-hour urine protein to detect significant proteinuria in PE.

Combining 18F-DOPA PET and MRI with perfusion-weighted imaging improves delineation of high-grade subregions in enhancing and non-enhancing gliomas prior treatment: a biopsy-controlled study.

We aimed to compare spatial extent of high-grade subregions detected with combined [18F]-dihydroxyphenylalanine (18F-DOPA) PET and MRI to the one provided by advanced multimodal MRI alone including Contrast-enhanced (CE) and Perfusion weighted imaging (PWI). Then, we compared the accuracy between imaging modalities, in a per biopsy analysis. Participants with suspected diffuse glioma were prospectively included between June 2018 and September 2019. Volumes of high-grade subregions were delineated respectively on 18F-DOPA PET and MRI (CE and PWI). Up to three per-surgical neuronavigation-guided biopsies were performed per patient. Thirty-eight biopsy samples from sixteen participants were analyzed. Six participants (38%) had grade IV IDH wild-type glioblastoma, six (38%) had grade III IDH-mutated astrocytoma and four (24%) had grade II IDH-mutated gliomas. Three patients had intratumoral heterogeneity with coexisting high- and low-grade tumor subregions. High-grade volumes determined with combined 18F-DOPA PET/MRI (median of 1.7 [interquartile range (IQR) 0.0, 19.1] mL) were larger than with multimodal MRI alone (median 1.3 [IQR 0.0, 12.8] mL) with low overlap (median Dice's coefficient 0.24 [IQR 0.08, 0.59]). Delineation volumes were substantially increased in five (31%) patients. In a per biopsy analysis, combined 18F-DOPA PET/MRI detected high-grade subregions with an accuracy of 58% compared to 42% (p = 0.03) with CE MRI alone and 50% (p = 0.25) using multimodal MRI (CE + PWI). The addition of 18F-DOPA PET to multimodal MRI (CE and PWI) enlarged the delineation volumes and enhanced overall accuracy for detection of high-grade subregions. Thus, combining 18F-DOPA with advanced MRI may improve treatment planning in newly diagnosed gliomas.

Taguchi array optimization of the reverse transcription loop-mediated isothermal amplification (RT-LAMP) assay for sensitive and rapid detection of dengue virus serotype 2.

Serotype 2 of dengue virus (DENV-2) is the most prevalent cause of dengue fevers. In this study, the C-prM gene was used for specific detection of DENV-2 by RT-LAMP assay. The RT-LAMP assay was optimized using the Taguchi design of experiments. The efficiency of the assay in such optimal conditions resulted in 100% sensitivity, 100% specificity, and 100% overall accuracy for detection of 4 copies/μL of the genome of DENV-2. In addition, the detection of 2 copies/μL of the genome of DENV-2 was feasible, although the sensitivity was 50%. Considering the importance of the specific detection of the dengue virus serotypes, the cost-effective RT-LAMP approach can be used for rapid, specific, and sensitive detection of DENV-2. RT-LAMP, as a cost-effective method, was optimized using Taguchi array approach for specific and rapid detection of DENV-2. Such methods can facilitate the diagnosis procedure in remote regions.

Comparative study of cardiac magnetic resonance imaging and transesophageal echocardiography in detection of left atrial and left atrial appendage thrombi

Abstract Funding Acknowledgements Type of funding sources: None. Background Left atrial (LA) thrombi are the most common intracardiac masses. Left atrial appendage (LAA) is the most common site for thrombus formation in patients with atrial fibrillation (AF), and in patients with valvular lesions. Transesophageal echocardiography (TEE) has been the conventional investigation for thrombus detection. However, because TEE is a semi-invasive technique; a completely non-invasive technique would be of great concern. Purpose The aim of this study is to evaluate the feasibility and diagnostic performance of cardiac magnetic resonance imaging (CMR), for the assessment of thrombi in the left atrium and left atrial appendage (LA/ LAA). Also whether CMR is comparable to TEE in identification and measurement of thrombus size. Methods We studied 43 patients who were diagnosed to have LA/LAA thrombus, or highly suspected thrombus by TEE. They underwent multisequence CMR for assessment of thrombus detection; within 7 days of TEE performance. Data collected from CMR study were statistically analyzed to evaluate for sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) and overall accuracy of detection of LA/LAA thrombus using TEE as the gold standard. Also agreement between both imaging techniques was assessed using kappa agreement coefficient. We conducted a questionnaire where 10 questions were asked to every patient in the study. It aimed to obtain an idea about patients’ opinion of both tests. Results During the study period of 13 months, 43 patients were assessed. Twenty one patients had AF and 22 patients were in sinus rhythm. The median CHA2DS2VASc score of AF patients was 2.52 ± 1.12, and 76.7% of patients were undergoing anticoagulation therapy. In all subjects, the LAA was readily visualized with CMR. When evaluating the diagnostic performance of CMR results revealed overall sensitivity, specificity, PPV, NPV and accuracy of 97.44%, 75%, 97.44%, 75% and 95.35% respectively. Results of questionnaire was statistically significant p value yielding good overall opinion for the sake of CMR. There was an additional role for CMR regarding tissue characterization; where CMR detected 3 cases, diagnosed by TEE as thrombus, 1 case revealed Libman-Sacks Endocarditis and 2 cases revealed fibroelastoma. This made a paradigm shift in patient management. Conclusion CMR is a noninvasive, feasible and comparable modality for thrombus detection in the LA and LAA and could be a reasonable, more comfortable alternative to TEE.

A clustering method for inter-annual NDVI time series

ABSTRACT Inter-annual Normalized Difference Vegetation Index (NDVI) time series has been applied to change detection in many fields. Research on change detection in inter-annual NDVI time series always requires several related parameters or sample training, which may limit the transferability of the methods to some extent. In this letter, we try to develop a shape-based clustering to analyse the inter-annual NDVI time series by considering the shapes of time series. The method adopts shape-based distance based on cross-correlation analysis to measure the distance between time series and uses shape-based averaging method named Dynamic Time Warping Barycentre Averaging to get the cluster centroids. Through experimental analysis, the shape-based clustering can well gather the NDVI time series with similar trends into a cluster. Also, the shape-based clustering is proven to be reliable with 94.4% overall accuracy for detection of vegetation disturbance.

IYOLO: Multi-scale infrared target detection method based on bidirectional feature fusion

Convolutional neural networks are advanced computer vision solution for various tasks, and have made considerable progress. In order to detect infrared targets with details and texture features, we propose a detection method based on YOLOv3. First, we propose a bidirectional feature fusion structure called Improve-FPN(Im-FPN), which allows fast and efficient multi-scale feature fusion. Second, we use the Focal Loss(FL) loss function to balance the problem of sample imbalance. Based on these optimizations, an Infrared-YOLO (IYOLO) target detection model for infrared target is constructed. On the self-constructed infrared dataset: Infrared-VOC, the mAP of IYOLO can achieve 77.1%, which is 4.0% higher than YOLOv3, and the detection speed can reach 55.6FPS. Experimental results show that this method can improve the overall accuracy of detection while ensuring real-time detection.

Identification of glottal instants using electroglottographic signal for vulnerable cases of voicing.

Robust detection of glottal instants is essential for various speech and biomedical applications. Glottal closing and glottal opening are two crucial instants/epochs of a glottal cycle. The first-order derivative of the Electroglottographic (EGG) signal demonstrates important peaks at those locations for standard voicing, but the detection of glottal instants becomes erroneous when the peak to peak amplitude of the EGG signal is very low, irregular and unpredictable. In this work, a new efficient method is proposed for identification of glottal instants from the EGG signals including the segments of the signals where the signals are feeble with irregular periodicity. The overall accuracy of detection will be enhanced by identifying the glottal instants for the whole part of the signal including the vulnerable segments of signal. As the phase of a signal is uniform in nature, the phase information of the EGG signal has been explored to detect glottal instants accurately. Under low strength of the EGG signal, the proposed method remarkably has better performance compared to the existing instants detection methods and for pathological EGG signal, the detection accuracy of glottal instants is better than other existing methods.

Ultrasound based evaluation of hepatic steatosis and fibrosis in hepatitis C non-responders.

To determine the accuracy of ultrasound in the diagnosis and grading of steatosis and fibrosis in Hepatitis C (HCV) patients not responding to ribavarin-interferon therapy. A cross-sectional, analytical study. Radiology Department, Civil Hospital, Karachi, from March 2008 to August 2010. Patients with positive HCV RNA despite 24 weeks ribavarin-interferon therapy (non-responders) were subjected to ultrasound and biopsy prior to institution of pegylated interferon therapy for detection and grading of steatosis and fibrosis. Using histopathology as the gold standard, sensitivity, specificity, negative and positive predictive values for ultrasound were determined. The sensitivity of ultrasound for hepatic steatosis was 90.9% for no steatosis (NS), 100% for moderate and gross steatosis and 84.4% for mild steatosis with 100% specificity. The sensitivity for fibrosis was 25% for no fibrosis, 100% for mild fibrosis, 89.74% for moderate fibrosis and 100% for gross fibrosis. The overall accuracy for detection of steatosis was 95.39% and that for fibrosis was 98.02%. Hepatic vein showed increased dampening of flow with advancing grades of steatosis and fibrosis. Ultrasound has a high accuracy in the diagnosis and grading of steatosis and fibrosis in HCV nonresponders. Mild fibrosis may confound the diagnosis of mild steatosis.

Rational use of 18F-FDG PET/CT in patients with advanced cutaneous melanoma: A systematic review.

18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is increasingly used in patients with advanced melanoma. Immune checkpoint inhibitors and BRAF/MEK-targeted therapy have transformed the therapeutic landscape of metastatic melanoma. Consequently, a need for markers predicting (early) response to treatment and for monitoring treatment (toxicity) has arisen. This systematic review appraises the current literature evidence for rational use of 18F-FDG PET/CT scans in staging, clinical decision-making, treatment monitoring and follow-up in advanced melanoma. 18F-FDG PET/CT has high overall accuracy for detection of distant metastases and is, combined with cerebral MRI, the preferred imaging strategy for staging metastatic melanoma. In contrast, strong evidence supporting the standard use of 18F-FDG PET/CT for predicting and monitoring therapy response and toxicity is currently lacking. Essential for determining the position of 18F-FDG PET/CT during treatment course in advanced melanoma are well-designed studies with standardized scanning protocols, incorporation of clinical parameters and comparison with contrast-enhanced CT alone.

Information-Guided Flame Detection Based on Faster R-CNN

Due to the diversity of the shape and texture of flame, and interference objects that similar to flame in color, detecting the position of flame from images is a difficult task. To enable generic object detection methods to achieve better performance in flame detection tasks, a color-guided anchoring strategy is proposed that uses color features of the flame to limit the location of the anchor. To solve the problem of high false alarm rate when directly using generic object detection methods in flame detection, a global information-guided flame detection method is proposed, this strategy uses a parallel network to generate image global information. We use these two methods to improve Faster R-CNN (Regions with Convolutional Neural Network features) to perform the fire detection process in a guided manner. Experiments on the BoWFire dataset show that our method improved detection speed by 10.1% compared with the original Faster R-CNN. In addition, the false alarm rate is decreased by 21.5%, and the overall accuracy of detection is increased by 9.3%. Experiments on PascalVOC and Corsician datasets further demonstrate the robustness of the proposed methods.

Inside the USCAP Journals

Inside the USCAP Journals

P6182Combination of a new iterative reconstruction technique with low tube voltage and high tube current has important role of detection of late enhancement on 320 slice CT

Abstract Background New iterative reconstruction tecniques, including Adaptive Iterative Dose Reduction 3D (AIDR 3D) and Forward Projected Model-based Iterative Reconstruction SoluTion (FIRST), have been recently available on new generation 320 slice CT, and they can provide high-quality CT images. Purpose The aim of this study was to evaluate the diagnostic performance of detection of abnormal late enhancement (LE) in left ventricular (LV) myocardium (LVM) using 320-slice CT with new iterative reconstruction techiniques, AIDR 3D (Figure A) and FIRST (Figure B). Methods A total of 100 patients who were suspected of having various myocardial diseases and underwent late phase acquisition both on cardiac CT and CMR within 3 months were analyzed. The first 50 consecutive patients (Group 1) underwent 320-slice CT with AIDR 3D, 120 Kv tube voltage, 519±71 mA tube current. The next 50 consecutive patients (Group 2) underwent 320-slice CT with FIRST, 80 or 100Kv tube voltage, 803±20 mA tube current. We compared diagnostic accuracy of CT for detection of LE in LVM against that of CMR (the gold standard) in between the 2 groups. Results On patient-by-patient analysis, sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and overall accuracy for detection of LE on CT vs CMR were 87, 95, 96, 82, and 90% in Group 1, and 97, 83, 91, 88, and 90% in Group 2. There were no significant difference of diagnostic accuracy on patient-by-patient analysis in between the 2 groups (Figure C). However, on a segment-by-segment analysis (using 17 American Heart Association LV segment model), these values for detection of LE on CT vs CMR were 60, 95, 73, 91, and 88% in Group 1, and 85, 95, 86, 95, and 93% in Group 2. Sensitivity, PPV, NPV and overall accuracy were significantly higher in Group 2 than in Group 1 (all P&lt;0.01) (Figure D). Conclusions Diagnostic accuracy of detection of LE in LVM on CT combining low tube voltage and high tube current acquisition on a new generation 320-slice CT with FIRST was superior to 320-slice CT with AIDR 3D. Acknowledgement/Funding TSUCHIYA MEMORIAL MEDICAL FOUNDATION

COMPARISON BETWEEN THE ROLE OF TRANSABDOMINAL ULTRASOUND VERSUS TRANSVAGINAL ULTRASOUND IN EVALUATION OF PLACENTAL INVASION IN CASES OF PLACENTA PREVIA ANTERIOR WALL WITH PREVIOUS UTERINE SCAR

Background: Accurate antenatal diagnosis of an abnormally invasive placenta (AIP), allowing multidisciplinary management at the time of delivery, has been shown to improve maternal and fetal outcomes. Placenta previa and previous cesarean section are the two most important known risk factors for AIP. Objective: Comparison between the role of transabdominal ultrasound (TAS) versus transvaginal ultrasound (TVS) in evaluation of placental invasion in cases of placenta previa anterior wall with previous uterine scar applying the unified ultrasonography descriptors suggested by the European working group on abnormally invasive placenta EW-AIP. Patients and Methods: Fifty pregnant women with persistent placenta previa (after 28 weeks’ gestation) were prospectively enrolled into this study. Both transabdominal and transvaginal ultrasound were performed by two different operators who were blinded to the results of each other. The placenta was studied as regarding the exact localization and the unified descriptors were applied and evaluated by TAS and TVS. The ultrasound findings were analyzed with reference to the final diagnosis made during cesarean delivery and histopathological examination. Results: Abnormally invasive placenta and its variants were found in 43 patients at the time of cesarean delivery, and were later confirmed by histopathological evaluation. As regards Grey-scale criteria; the accuracy of detection of the loss of the retro placental clear zone was 76% by TVS and 54% by TAS, while that of the abnormal placental lacunae was 92 % by TVS and 88% by TAS. Myometrial Thinning detection accuracy was 66% by TVS and 72% by TAS, and Doppler assessment showed that the accuracy of detection of the uterovesical hyper vascularity was 84% by TVS and 88% by TAS. The detection of bridging vessels was 76% by TVS and 75% by TAS. The overall accuracy of detection of TAS was 91%, whereas that of TVS was about 97.1%. Conclusion: Both transabdominal and trasnvaginal ultrasound were complementary to each other with an upper hand to transvaginal ultrasound, with the safety of TVS being confirmed. The unified descriptors were found to be reliable in accurate diagnosis.

RIMS: A Real-time and Intelligent Monitoring System for live-broadcasting platforms

Personal live shows on Internet streaming platforms currently are blooming as one of the most popular applications on mobile phones and especially attracting millions of young generation users. The content supervision on live streaming platforms, in which there are thousands or hundreds of show rooms for performing and chatting synchronously, is a major concern with the development of this new service. Traditional image captures and real-time content analysis experience huge difficulties such as processing delay, data overwhelming, and matching overhead. In this paper, we propose a comprehensive method to monitor real-time live stream and to identify illegal or unchartered live misbehaviors intelligently based on various proposed aspects instead of image analysis only. The proposed system called RIMS makes use of several novel indicators on show room status rather than analyzing images solely to support real-time requirements. Three detecting techniques are adopted: self-adaptive threshold-based abnormal traffic detection, sensitive Danmaku comment perception, and frame difference analysis. RIMS can detect dramatically increasing of user number in a show room, filter sensitive words in Danmaku, and capture segmentation of video scenes by frame difference analysis. We deploy our system to monitor a typical live- broadcasting platform called panda.tv, and overall accuracy of detection via three indicators reaches 90.1%. The application of RIMS can change current supervison methods on live platforms that they totally rely on real-time manual review or after the event check. The key techniques in RIMS can also be widely employed in many other mobile applications in edge computing such as video surveillance in Internet of Things and mobile short video sharing.

P30.07: Prospective evaluation of ultrasound implementation into preoperative evaluation of cervical cancer patients

To assess if ultrasound can be widely implemented in cervical cancer staging - to assess the accuracy of ultrasound done by less experienced sonographer after training. Multicentre prospective study. After theoretical and practical education, 18 sonographers in 11 centres in one country assessed cervical cancer patients per standardised protocol. Ultrasound findings were compared to histopathological examination. The accuracy of ultrasound was assessed for detection of residual tumour, measurement of tumour volume and false negativity for detection of parametrial involvement. 228 consecutive patients with both ultrasound and pathological reports were included into final analysis. The overall accuracy for detection of residual tumour was 93%, for measurement of tumour volume (agreement between ultrasound volumometry and histopathology) 89% and false negativity of parametrial involvement was 2%. There was no difference in the accuracy between referral (teaching) centre and other centres involved. Ultrasound is a feasible method to be used for local staging of cervical cancer patients. With appropriate short-term education of sonographer, it can be implemented into the standard clinical management.

MiR-205 as a promising biomarker in the diagnosis and prognosis of lung cancer.

MicroRNA-205 (miR-205) was revealed as a novel diagnostic and prognostic biomarker for lung cancer, but the results in the published papers were inconsistent. This meta-analysis aimed to investigate the diagnostic and prognostic roles of miR-205 in patients with lung cancer. Totally, 16 eligible articles were included, among which 10 articles investigated the diagnostic value of miR-205, 5 articles examined its prognostic values, and 1 article studied both diagnostic and prognostic values. For the diagnostic meta-analysis, the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the overall area under the curve of miR-205 for patients with lung cancer were 0.88 (95% CI = 0.78 – 0.94), 0.78 (95% CI = 0.66 – 0.86), 4.00 (95% CI = 2.47 – 6.49), 0.16 (95% CI = 0.08 – 0.30), 25.86 (95% CI = 9.29 – 71.95), and 0.90 (95% CI = 0.87 – 0.92), respectively, indicating that miR-205 is a useful biomarker for diagnostic of lung cancer. The subgroup analysis further demonstrated that miR-205 had an excellent overall accuracy for detection with tissue samples compare with blood samples. For the prognostic meta-analysis, the pooled outcome of the disease-free survival and recurrence-free survival analyses revealed that increased miR-205 levels had a protective role in the prognosis of patients with lung cancer (pooled HR = 0.86, 95% CI: 0.78-0.96, z = 2.83, P = 0.005). In conclusion, miR-205 may be a promising biomarker for detection, predicting the recurrence of patients with lung cancer.